w.elsevier.com/locate/tsf

Thin Solid Films 515 (

Assembling of redox proteins on Au(111) surfaces: A scanning probe

microscopy investigation for application in bio-nanodevices

L. Andolfi, A.R. Bizzarri, S. Cannistraro *

Biophysics and Nanoscience Centre, INFM-CNISM, Dipartimento di Scienze Ambientali, Universita della Tuscia, Viterbo, I-01100, Italy

Available online 19 January 2006

Abstract

The morphology and conductive properties of azurin molecules, chemically attached to sulfhydryl terminated alkanethiol monolayer assembled

on Au(111) surface, are mapped at single molecule level and compared with those observed for the same molecule immobilised on bare Au(111).

High-resolution Tapping Mode Atomic Force Microscopy shows that the protein molecules immobilised on modified gold, better reproduces the

crystallographic height of the protein, than that immobilised on bare gold. Such a height recovering is also found in the Scanning Tunnelling

Microscopy images. Consistently, a good tunnelling conduction of azurins on the modified gold electrode is demonstrated by Tunnelling

Spectroscopy. Cyclic voltammetry measurements show, in addition, that the redox activity of azurin molecules covalently immobilised on

sulfhydryl functionalised Au(111) surface is retained. These results are discussed in connection with possible use of this linker in the assembling

of nano-hybrid systems.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Scanning tunnelling microscopy; Atomic force microscopy; Azurin; Conduction; Sulfhydryl-terminated Au(111)

1. Introduction

The structural organization and the electron transfer (ET)

properties of redox metalloproteins assembled on metallic

surfaces are of central interest in the new area of nano-

biotechnology. This field aims at creating devices of enhanced

sensitivity that could be useful in biomedical and environmen-

tal research and drug discovery [1–3]. ET proteins offer a

number of advantages in the construction of nano-biosensors

due to their nanosized structure and their redox activity, which

can be suitably tuned and electrochemically monitored [4,5].

The integration of ET proteins with a metallic electrode, with

the possibility of establishing a good electrical contact among

them, is a crucial issue in nano-biosensing. The ability of

revealing an electrical signal, and therefore the sensitivity and

reliability of these nano-biosensors, is strongly dependent on

how protein immobilisation on the metal surface is achieved

[6]. Obviously, it is also very important for the assembling of

the ET proteins to ensure a preserved protein structure and

function. An efficient electrical communication can be

0040-6090/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.tsf.2005.12.115

* Corresponding author. Tel.: +39 0761 357136; fax: +39 0761 357179.

E-mail address: cannistr@unitus.it (S. Cannistraro).

achieved by covalently linking the protein directly to the metal

electrode. For such a purpose, thiols and/or disulfide groups

present (or introduced by site-directed mutagenesis) in the

protein molecules can be exploited, due to their ability to form

stable bonds with the gold electrodes [7–13]. Direct protein

adsorption on metallic surface provides the advantage to keep

distances between the redox centre and the electrode surface

within the range at which significant ET rates can occur [14].

Morphological and functional properties of some redox

proteins assembled on bare gold have been studied by

voltammetry, ellipsometry, X-ray photoelectron spectroscopy

and mapped to single molecule level by scanning probe

microscopy [7–18]. These techniques have shown that the

assembled redox proteins can retain, in some cases, their

function upon immobilization on gold [7,9,11,15,16]. In other

cases, a partial protein unfolding has been observed by height

analysis in Atomic Force Microscopy (AFM), in conjunction

with ellipsometry studies on protein monolayers [7,8,17,18]. It

has been also observed that some redox proteins, directly

attached to a metal, do not give any electrochemical signal

[19–21]. These results could be related to the occurrence of an

extensive protein interaction with the metallic surface, which

can eventually lead to a loss of protein native conformation.

2006) 212 – 219

ww

L. Andolfi et al. / Thin Solid Films 515 (2006) 212–219 213

However, Scanning Tunnelling Microscopy (STM) operating

in constant current mode almost always underestimates the

physical height of the assembled proteins on Au(111) surface

with respect to other techniques (crystallography, AFM and

ellipsometry) [11,13,15,16,22–25].

A more gentle linking of the proteins to metallic electrodes

can be attained by chemically modifying the metal surface by

assembling a monolayer of short alkane molecules, on the top

of which the proteins can be anchored [26–29]. A variety of

small organic molecules can be used to create sulfhydryl-

terminated alkanethiol monolayer on a gold surface, which can

react with thiol groups of the biomolecules leading to a well

defined protein immobilization [30,31]. It should be taken into

account that a suitable linker should not significantly affect the

ET rate due to its length.

In this work, we used an amine-termined monolayer formed

by self-assembling cysteamine molecules on Au(111), which

was reacted then with the heterobifunctional linker N-succini-

midyl-S-acetylthiopropionate to obtain a sulfhydryl surface

[32,33]. On this sulfhydryl functionalised Au(111) surfaces we

tethered azurin (AZ) molecules. AZ is a small redox copper

protein, which bears an exposed disulfide group, located in a

region opposite to the redox site and suitable for a covalent

anchoring on both bare and thiol-terminated gold surface. As

schematically shown in Fig. 1A, AZ molecules assembled on

the modified Au(111) are expected to exhibit a molecular

orientation similar to that obtained on bare gold (see Fig. 1B),

but with the protein residues sheltered from a direct strong

interaction with gold [34].

Fig. 1. Schematic view of AZ molecular orientation upon reaction of S–S

moiety with the sulfhydryl terminated Au(111) surface (A); AZ molecular

orientation on bare Au(111) when protein is anchored via S–S group (B).

The morphological and conductive properties of individual

AZ proteins assembled on sulfhydryl-terminated monolayer

have been investigated by high-resolution Tapping Mode AFM

(TMAFM), STM and Scanning Tunnelling Spectroscopy

(STS), and compared with those obtained for AZ on bare

Au(111). The vertical size of AZ molecules immobilised on

thiol-modified Au(111), evaluated by TMAFM, closely

matches the crystallographic value [35], while on bare gold a

reduced AZ height has been generally obtained. STM images

of AZ molecules on thiol functionalised Au(111) show single-

molecule structures with a height notably enhanced as

compared to that obtained for AZ, and in general for other

biomolecules, adsorbed on gold. Moreover, STS data reveal a

good tunnelling conduction for AZ immobilised on sulfhydryl

terminated gold.

Analogously to what was found for the AZ proteins

anchored directly on gold, the redox activity of AZ molecules

on chemically modified gold surface is retained, as demon-

strated by cyclic voltammetry (CV) measurements.

2. Experimental

AZ, cysteamine and N-succinimidyl-S-acetylthiopropionate

(SATP) have been purchased from Sigma Chemical Co., and

used without further purification. Gold substrates (Arrandee)

consist of a vacuum evaporated thin gold film (thickness

200 nm) on borosilicate glass. They have been annealed with a

butane flame to obtain re-crystallized Au(111) terraces. The

quality of the annealed gold surface was assessed by STM,

which showed atomically flat (111) terraces over hundreds of

nanometers.

The molecules were adsorbed on Au(111) surface by

directly incubating the annealed substrates with AZ protein

solution (3.5 AM in 50 mM NH4Ac pH 4.8) at 4 -C for times

ranging between 30 min and few hours.

The Au(111) surface functionalization was made by

immersing freshly annealed substrates into a 1 mM ethanolic

solution of cysteamine for 24 h. SATP (20 mM, 10% DMSO

and 90% PBS pH 7.0) was reacted with the cysteamine

monolayer for 2 h. The protecting group of the sulfhydryl was

removed by exposing the monolayer to a solution of 0.5 M of

hydroxylamine in 50 mM PBS pH 7, 25 mM EDTA, and

50 mM DTT for 20 min. The sulfhydryl surface was then

reacted with AZ solution (20 AM in 50 mM NH4Ac pH 4.8)

for 1 h at room temperature. Samples were then rinsed with

ultrapure water and blown dry with pure nitrogen.

TMAFM measurements were performed with a Nanoscope

IIIa/Multimode, Digital Instruments equipped with a 12-Amscanner operating in tapping mode in ultra-pure water

(18.2MV cm). Silicon probes (Digital Instruments), 100 or

200 Am long, with nominal radius of curvature less than

about 20 nm and spring constants of 0.15 and 0.57 N/m,

respectively, were used. Resonance peaks in the frequency

response of the cantilever were chosen in the range of 8–

30 kHz. Free oscillation of the cantilever was set to have

root-mean-square amplitude corresponding to 10 nm. In each

measurement, the set point was adjusted before scanning, to

C-N-OH

H3C

O

NO

N OO

Au SNH2

+ S

O OO

O (SATP)

CH3

OH

N SAu S

O O

HCH3

NAu S

HSH

O

S

S

Azurin

N SAu S

O O

HCH3 + H2N-OH

(Hydroxylamine)

(Cysteamine)

NAu S

HS

O

S

HS

Azurin

Au + HSNH2 redox

Au SNH2 + 1/2 H2

A

B

C

D

E

F

Cysteamine monolayer assembling on Au(111)

Thiolation with N-succinimidyl S-acetylthiopropionate

Sulfhydryl deprotection with Hydroxylamine

Protein adsorption via S-S bond

Fig. 2. Surface reaction scheme illustrating the steps involved in formation of chemically modified Au(111) surface. (A) self-assembling of cysteamine on gold via

SH group; (B–C) reaction of the heterobifunctional linker SATP with the cysteamine monolayer forming an amide bond; (D) deprotection of the sulfhydryl by

removing the acetyl protecting group; (E–F) the sulfhydryl active group reaction with the S–S group of AZ.

L. Andolfi et al. / Thin Solid Films 515 (2006) 212–219214

minimise the force between the tip and the sample. The

height related to the z-piezo and the curvature radius of the

tips were calibrated by using 5 nm gold colloids deposited on

a glass slide coated with (3 mercaptopropyl)-trimethoxysilane

[36].

A Picoscan system (Molecular Imaging) equipped with a

10 Am scanner with a final preamplifier sensitivity of 1 nA/V

was used for STM and STS measurements. STM tips were

prepared by mechanically cutting Pt/Ir wires (Goodfellow). For

STS experiments, Current–voltage (I –V) curves were obtained

by setting the gap between the STM tip and the protein at a

tunnelling current of 50 pA and bias of �1 V. Then the

feedback was disengaged and the current was monitored as the

substrate potential is swept over T1 V. Every single sweep was

collected in 0.01 s.

Cyclic voltammetry was performed with a PicoSTAT

bipotentiostat (Molecular Imaging Co.). The electrochemical

cell housed two Pt wires as counter and pseudo-reference

electrodes and was filled with 150 Al of 50 mM NH4Ac pH 4.8.

The potential of Pt wire was calibrated against a standard

Fig. 3. Representative TMAFM images acquired on AZ molecules directly adsorbed on Au(111) surface (A), and on AZ adsorbed on cysteamine-SATP monolayer

(B). Cross section profiles of the molecules indicated by the white arrows are reported in the lateral panels.

L. Andolfi et al. / Thin Solid Films 515 (2006) 212–219 215

calomel electrode (SCE). All potentials are then quoted relative

to SCE.

3. Results and discussion

The thiol-functionalization of Au(111) surface, for oriented

AZ immobilization, was formed according to the reaction

scheme depicted in Fig. 2. The cysteamine molecules self-

assemble on Au(111) via their thiol group generating an amine

terminated monolayer (Fig. 2A). These groups can then react

with the heterobifunctional linker SATP (Fig. 2B), which on

one hand forms a stable amide linkage with the amine group,

while on the other exposes an acetyl group protecting a

sulfhydryl group (Fig. 2C). By treating the monolayer with

hydroxylamine the acetyl group is removed revealing an active

sulfhydryl surface (Fig. 2D). The sulfhydryl groups are then

exposed to react with the disulfide bond of AZ (Fig. 2E,F).

This attachment chemistry has been confirmed by spectro-

scopic studies [32,33], and produces an ordered uniform

monolayer suitable for proteins tethering and for nanoscale

studies.

The morphology of AZ molecules assembled on bare

Au(111) and on the cysteamine-SATP monolayer assembled

on gold was characterized by TMAFM. This technique, while

on one hand gives minor information about the lateral

molecular size owing to the broadening effects introduced by

the tip size, on the other accurately estimates the vertical

dimension of the biomolecules over the substrates. Fig. 3A

L. Andolfi et al. / Thin Solid Films 515 (2006) 212–219216

shows an AFM image of AZ molecules directly adsorbed on

Au(111), via the S–S group as widely demonstrated

[7,8,11,17]. In this image, single molecules are clearly detected

on a quite smooth surface, which display a root-mean-square

roughness (RMS) of 0.13T0.01 nm. The protein height with

respect to the Au(111) substrate was evaluated by a cross

section analysis on individual molecules (see cross section

profile of Fig. 3A). Over a collection of 100 molecules, we

obtained a gaussian distribution with a mean value of 1.7 nm

and a standard deviation of 0.5 nm, consistent with previous

studies [7,8,17]. This value appears to be lower than what was

expected by the crystallographic structure, if the AZ assem-

bling on gold, via S–S group, would occur in a standing up

arrangement as illustrated in Fig. 1B [7]. Such finding could be

likely associated with a strong AZ interaction with the noble

metal, that might either force the protein to adopt a lying down

configuration above the substrate or even to cause a partial

protein denaturation.

A representative AFM image acquired on AZ molecules

adsorbed on the cysteamine-SATP monolayer assembled on

Au(111) is shown in Fig. 3B. Over a surface background with a

roughness RMS=(0.40T0.04) nm, single AZ molecule are

well resolved as homogeneous globular shape structures. The

molecular vertical dimension of AZ on the monolayer is

evaluated by cross section analysis on individual molecules

(see height profile of Fig. 3B), also taking into account

contributions of the background roughness. The obtained

heights are plotted in the histogram of Fig. 4. The distribution

is centred at a mean value of 3.4 nm with a standard deviation

of 0.8 nm. This value, significantly higher than that observed

for AZ directly immobilised on Au(111), well matches the

vertical structure of the protein, as evaluated by crystallography

[35]. Such result indicates that the interactions between the

amino acid residues and the noble metal are screened upon the

binding of AZ (via its S–S bridge) with the thiol group of the

1.6 2.4 3.2 4.0 4.8 5.60

2

4

6

8

10

12

14

16

18

20

22

24

26

Occ

urre

nce

(num

ber

of m

olec

ules

)

Molecular height (nm)

Fig. 4. Statistical analysis of AZ molecular height on chemically modified

Au(111) surface, mean height value=3.4 nm and r =0.8 nm. Data are obtained

from individual cross section profiles over 100 molecules.

monolayer, and that under these conditions the protein may

adopt a standing up configuration on the substrate, with a three-

dimensional structure closer to its native form.

A further characterization of AZ molecules assembled on

bare and cysteamine-SATP modified Au(111) was performed

by STM, operating in constant current mode, where the

molecular height is registered as a function of the measured

current. The protein lateral dimension can be precisely

evaluated by STM, where tip is known to induce a minor

convolution with respect to AFM. Fig. 5A shows an STM

image of single AZ molecules adsorbed on Au(111); they are

stable upon repetitive scans and present a lateral size of

4.5T0.9 nm (see cross section profile of Fig. 5A), consistently

with other works [7,8,11,17]. The molecular height, on the

contrary, is found to be 0.5T0.1 nm, appearing significantly

lower than the physical height evaluated by crystallographic

studies [35]. This is, however, a general characteristic of STM

images obtained on biomolecules self-assembled on conductive

substrates [11,13,16,22–25], and it is very likely related to the

low conductivity of biomolecules. Such an aspect has been

deeply addressed in a previous work, where it is shown that the

real vertical dimension is recovered by STM only when

uniformly metallic nanoparticles deposited on Au(111) sub-

strates are imaged [25]. For redox proteins adsorbed on gold,

variations of the molecular height have been observed by

performing STM under electrochemical control, which has

indicated that, in some cases, tunnelling current flow through

the redox protein can be properly modulated upon tuning the

substrate potential with the respect to the redox potential of the

protein [13,23].

Surprisingly, we find that the STM height of the AZ

molecules assembled on the cysteamine-SATP monolayer is

1.8 nm with a standard deviation of 0.4 nm (see Fig. 5B). Such

enhanced molecular height is put into evidence by a

representative height profile shown in the lateral panel of

Fig. 5B. The protein lateral size, obtained in the STM images,

is 3.7 nm with a standard deviation of 0.8 nm, consistent with

that found for AZ on bare gold. The vertical and lateral

dimensions of the protein molecules appear to be reproducible

after repetitive scans. The significant increment in the

measured molecular height can result from a more efficient

electron tunnelling between the tip and the substrate through

the protein when covalent immobilization is achieved by the

cysteamine-SATP linker.

To get additional information about the conduction of

single AZ immobilised on cysteamine-SATP monolayer, I –V

curves were registered by STS. In these measurements the tip

was positioned on top of a single protein, the feedback loop

was temporarily disengaged and the tunnelling current was

monitored as the sample bias was ramped in the range of

T1 V. Each single I –V curve, acquired on a single protein,

consists of the average over 10 consecutive bias sweeps.

These measurements were repeated on several molecules and

were averaged over 100 bias sweeps. The resulting curve for

AZ assembled on cysteamine-SATP, compared with those of

the bare and cysteamine-SATP modified Au(111), are shown

in Fig. 6.

Fig. 5. Constant current STM images of AZ molecules immobilised on Au(111) (A) and of AZ anchored on cysteamine-SATP monolayer assembled on gold (B). The

cross section profiles of the molecules (indicated by the white arrows) are shown in the lateral panels. Tunnelling current 50 pA and voltage bias �1 V; scan rate

3.0 Hz.

L. Andolfi et al. / Thin Solid Films 515 (2006) 212–219 217

The three curves appear to be superimposed in the negative

part of the I –V spectrum, while at positive bias (about +0.9 V)

we can observe that the monolayer deposited on gold registers

a small reduction of the current response. However, for AZ

bound on the monolayer the current response increases

approaching that obtained for bare gold. A slight asymmetry

for AZ assembled on the functionalised gold can be noticed,

which, in any cases, is not as pronounced as that observed for

AZ on bare Au(111) [7,17]. We remark that, although the STS

spectra show that the AZ tethered on cysteamine-SATP

monolayer are equally conductive within a range of +1 and

�1 V, we found some difficulties when STM imaging of these

sample was performed at positive biases (between +0.2 and

+1 V). In this case, the imaging appears strongly disturbed and

single molecule could not be detected. This behaviour is

generally indicative of a strong interaction between the STM

tip and the imaged sample. Conversely, such phenomenon is

not observed for AZ molecules directly anchored on gold,

which can be clearly imaged at positive and negative bias

values, without considerable variations in molecular height.

Although a good tunnelling conduction is revealed, the

discrepancy between spectroscopy and imaging observed both

Fig. 7. Voltammogram of AZ immobilised on SATP-cysteamine monolayer

recorded at a scan rate of 100 mV/s in 50 mM NH4Ac pH 4.8. The inset shows

the change in oxidation current with increasing scan rate.

L. Andolfi et al. / Thin Solid Films 515 (2006) 212–219218

for AZ immobilised on bare and modified gold is presently not

clear and requires further investigations.

Finally, the functionality of immobilised AZ on modified

gold electrodes was investigated by CV in which the faradaic

current was measured as function of the substrate potential. No

redox response was observed for the SATP-cysteamine

functionalised Au (111) substrate in 50 mM ammonium acetate

pH 4.8. A current response was obtained after overnight

incubation of the activated monolayer with AZ solution. The

voltammogram of Fig. 7 shows a pair of peaks corresponding

to the oxidation and reduction peak of the protein on the

cysteamine-SATP monolayer. The voltammetric response is

stable up to few hours of measurements. The formal redox

potential (E1 / 2), calculated as E1 / 2= (Epa+Epc) / 2, is 280T20 mV, which appears to be shifted to more positive values

than that reported for AZ directly anchored on gold (165–

175 mV) [15]. The linear dependence of the peak current on the

scan rate is consistent with electroactive molecules being

confined to the surface (see inset of Fig. 7). The separation

between the anodic and the cathodic peaks DEp, is 170 mV (at a

scan rate of 100 mV/s) and it is dependent on the scan rate

showing a quasi-reversible kinetics. Moreover, the DEp value

obtained on themodified gold is greater than that of AZ adsorbed

on bare Au(111), indicating a slower electron transfer process.

The shift of the midpoint redox potential and the indication of a

slower ET are very likely the result of the cysteamine-SATP

monolayer, being interposed between the protein and the

Au(111) surface. An estimate of the surface coverage with

electroactive AZ molecules can be obtained from Eq. (1)

Ip mð Þ ¼ Nn2F2=4RT� �

m ð1Þ

where Ip is the peak current (anodic or cathodic), v is the voltage

scan rate, N is the number of redox-active sites on the surface, n

is the number of electrons transferred, F the Faraday constant, R

the gas constant and T the temperature. From the slope of Ipversus scan rate with n=1, we estimate a surface coverage of

2.1�1013 molecules cm�2. This value, in good agreement with

the expected coverage for a molecule with a lateral dimension of

Fig. 6. I –V curves recorded in ambient conditions on AZ molecules (open

circle), cysteamine-SATP monolayer (square) and Au(111) (solid line). The

engage tunnelling current and voltage bias are 50 pA and �1 V, respectively.

about 4 nm [15], confirms a high degree of structural retention of

the AZ proteins on the modified gold.

4. Conclusions

The present study indicates that AZ molecules can be firmly

and functionally assembled on both bare and suitably modified

Au(111) surfaces. The comparison of the two immobilization

strategies shows that the cysteamine-SATP monolayer inter-

posed between the gold surface and the AZ molecules, aids to

reduce the protein–metal interactions, resulting in a standing

up configuration of AZ molecules over the substrate with

protein structure closer to its native form. Strikingly, we found

that the sulfhydryl terminated alkanethiol monolayer is able to

facilitate the tunnelling current through the protein. Hence, the

cysteamine-SATP linker appears to be an effective way for

integrating the redox metalloproteins with a gold electrode, and

this represents an important result, especially in view of a nano-

biotechnology application of these proteins.

Acknowledgments

This work has been partially supported by the FIRB-MIUR

Project ‘‘Molecular Nanodevices’’ and a PRIN-MIUR 2004

project. L. Andolfi acknowledges the Research Grant MIUR

‘‘Rientro dei Cervelli’’.

References

[1] R. Freitas, Nanomedicine: Nanotechnology, Biology and Medicine,

vol. 1, 2005, p. 2.

[2] H. Klefenz, Eng. Life Sci. 4 (2004) 211.

[3] C.R. Lowe, Curr. Opin. Struck. Biol. 10 (2000) 428.

[4] I. Willner, B. Willner, Trends Biotech. 19 (2001) 222.

[5] G. Gilardi, A. Fantuzzi, Trends Biotech. 19 (2001) 468.

[6] E. Katz, V. Heleg-Shabtai, B. Willner, I. Willner, A.F. Buckmann,

Bioelectrochem. Bioenerg. 42 (1997) 95.

[7] B. Bonanni, D. Alliata, L. Andolfi, A.R. Bizzarri, S. Cannistraro, in: C.P.

Norris (Ed.), Surface Science Research Developments, Nova Science

Publishers, Inc., 2004.

[8] J.J. Davis, H.A.O. Hill, Chem. Commun. (2002) 393 .

L. Andolfi et al. / Thin Solid Films 515 (2006) 212–219 219

[9] H.A. Heering, F.G.M. Wiertz, C. Dekker, S. de Vries, J. Am. Chem. Soc.

126 (2004) 11103.

[10] L. Andolfi, S. Cannistraro, G.W. Canters, P. Facci, A.G. Ficca, I.M.C. Van

Amsterdam, M.Ph. Verbeet, Arch. Biochem. Biophys. 399 (2002) 81.

[11] Q. Chi, J. Zhang, J.U. Nielsen, E.P. Friis, I. Chorkendorff, G.W. Canters,

J.E.T. Andersen, J. Ulstrup, J. Am. Chem. Soc. 122 (2000) 4047.

[12] J. Hang, Q. Chi, T. Albrecht, A.M. Kuznestov, M. Grubb, A.G. Hansen,

H. Wackerbarth, A.C. Welinder, J. Ulstrup, Electrochim. Acta 50 (2005)

3143.

[13] L. Andolfi, B. Bonanni, G.W. Canters, M.Ph. Verbeet, S. Cannistraro,

Surf. Sci. 530 (2003) 181.

[14] R.A. Marcus, N. Sutin, Biochim. Biophys. Acta 811 (1985) 265.

[15] L. Andolfi, D. Bruce, S. Cannistraro, G.W. Canters, J.J. Davis, H.A.O.

Hill, J. Crozier, M.Ph. Verbeet, C.L. Wrathmell, Y. Astier, J. Electroanal.

Chem. 565 (2004) 21;

J.J. Davis, D. Bruce, G.W. Canters, J. Crozier, H.A.O. Hill, Chem.

Commun. (2003) 576.

[16] B. Bonanni, D. Alliata, A.R. Bizzarri, S. Cannistraro, Chem. Phys. Chem.

4 (2003) 1183.

[17] A.R. Bizzarri, L. Andolfi, M. Stchakovsky, S. Cannistraro, J. Nanotech-

nol. 1 (2005) 100.

[18] B. Schnyder, R. Koetz, D. Alliata, P. Facci, Surf. Int. Anal. 34 (2002) 40.

[19] D.E. Reed, F.M. Hawkridge, Anal. Chem. 59 (1987) 2334.

[20] F.A. Armstrong, Struct. Bond. 72 (1990) 137.

[21] Y. Zhou, T. Nagaoka, G. Zhu, Biophys. Chem. 79 (1999) 55.

[22] J.J. Davis, D.A. Morgan, C.L. Wrathmell, A. Zhao, IEE Proc.-Nanobio-

technol. 151 (2004) 37.

[23] P. Facci, D. Alliata, S. Cannistraro, Ultramicroscopy 89 (2001) 291.

[24] D. Losic, J.G. Shapter, J.J. Gooding, Langmuir 18 (2002) 5422.

[25] D. Alliata, L. Andolfi, S. Cannistraro, Ultramicroscopy 101 (2004) 231.

[26] Q. Chi, J. Zhang, J.E.T. Andersen, J. Ulstrup, J. Phys. Chem., B 105

(2001) 4669.

[27] M.J. Tarlov, E.F. Bowden, J. Am. Chem. Soc. 113 (1991) 1847.

[28] A.K. Gaigalas, G. Niaura, J. Colloid Interface Sci. 193 (1997) 60.

[29] O. Cavalleri, C. Natale, M.E. Stroppolo, A. Relini, E. Cosulich, S. Thea,

M. Novi, A. Gliozzi, Phys. Chem. Chem. Phys. 2 (2000) 4630.

[30] J. Zhang, H.E.M. Christensen, B.L. Ooi, J. Ulstrup, Langmuir 20 (2004)

10200.

[31] J. Zhang, A.C. Welinder, A.G. Hansen, H.E.M. Christensen, J. Ulstrup,

J. Phys. Chem., B 107 (2003) 12480.

[32] E.A. Smith, M.J. Wanat, Y. Cheng, S.V.P. Barreira, A.G. Frutos, R.M.

Corn, Langmuir 17 (2001) 2502.

[33] G.J. Wegner, H.J. Lee, G. Marriott, R.M. Corn, Anal. Chem. 75 (2003)

4740.

[34] X. Chen, R. Ferrigno, J. Yang, G.M. Whiteside, Langmuir 18 (2002)

7009.

[35] H. Nar, A. Messerschmidt, R. Huber, M. van de Kamp, G.W. Canters,

J. Mol. Biol. 218 (1991) 427.

[36] B. Bonanni, S. Cannistraro, J. Nanotechnol. 90106 (2005) 1.